Which of your channels actually make money
The same €12 dish leaves your kitchen identical every time. What it nets you depends entirely on which door it walks out of. A plate served at a dine-in table, the same plate handed over the counter as takeaway, and the same plate riding out on a courier all land you a completely different number — and most operators have never actually computed it.
That is the whole problem. You cannot decide what to do about a channel until you know what it earns you. So let us make the number visible.
The same dish, three different margins
Look at the bars. Dine-in is strong because almost nothing sits between the menu price and your pocket. Takeaway is a little thinner — packaging, a busier counter — but healthy. The marketplace bar is thin, and on a promo day it can dip below zero. The dish did not change. The channel did.
Here is why the gap is so wide. Marketplace commissions run 15–30%, but the effective cost reaches up to 40% once payment processing, marketing surcharges and promo cost-share are stacked on top. Set that against a typical restaurant net margin of about 8%, and the arithmetic is brutal: a 30% commission loses money on the order unless the menu price is inflated 15–30% to absorb it.
A 30% commission loses money on the order unless the menu price is inflated 15 to 30 percent to cover it.
Why the marketplace bar is so thin
The commission rate you quote from memory is never the real cost. It is the first line of a stack.
Base commission is predictable. What thins the bar is everything piled on afterward: payment processing, a marketing or visibility surcharge, and the promo cost-share you agreed to when you switched on "free delivery" and half-forgot. Each is small on its own. Together they carry a 15–30% headline up to an effective cost near 40% — and that is the number that eats an 8% margin whole.
Run the audit: a per-channel P&L
You do not need a finance team. You need one honest table, built from real orders over a real week. Take the same dish and walk it through each channel:
| Line | Dine-in | Takeaway | Marketplace |
|---|---|---|---|
| Menu price | €12.00 | €12.00 | €12.00 |
| Channel commission | €0.00 | €0.00 | −€3.60 |
| Payment processing | −€0.24 | −€0.24 | −€0.30 |
| Packaging | €0.00 | −€0.40 | −€0.55 |
| Marketing / promo cost-share | €0.00 | €0.00 | −€1.20 |
| Food cost (30%) | −€3.60 | −€3.60 | −€3.60 |
| Net before labour | €8.16 | €7.76 | €2.75 |
Same dish, same kitchen, three very different numbers. Now split each column by day-part and by promo, and the picture sharpens again: the marketplace order that nets €2.75 at a quiet 10pm — genuinely new demand you would not have had — is a different animal from the marketplace order that nets almost nothing at a full noon lunch, cannibalising a table you could have turned yourself.
Why nobody runs this — and what to fix
Understanding true profitability by channel is essential but rarely straightforward. The reason is structural: the average restaurant lists on about 2.5 delivery platforms, and each platform keeps its own back-office. When the numbers live in three or four systems that cannot be compared, nobody ever lines them up — and operators lose an estimated 20–30% of potential revenue to exactly this fragmentation, unable to see which channel is carrying the business and which is quietly draining it.
That decision only becomes possible when every order — dine-in, takeaway and every marketplace — lands in one queue and gets reconciled against the money actually received for it. One queue and one reconciliation is what makes the per-channel number visible in the first place.
The apps are not the enemy — they are demand you would not otherwise have. Not knowing which channel actually makes money is. Put every order on one screen and let analytics tell you the honest number, and the decision stops being a guess.